The proposed research is intended to evaluate treatment directed to misarticulating children and to predict treatment outcome. Two subject categories are employed: cleft palate and functional articulation disorder. The cleft palate studies have three purposes: (1) evaluate biofeedback training directed to velopharyngeal closure, (2) evaluate the influence of oronasal fistulas on speech and response to speech training, and (3) determine what speech gains can be achieved through training by individuals with velopharyngeal closure deficits. These studies measure articulation in single subject, repeated measures research designs. Video-endoscopic and pressure-flow instrumentation is used for biofeedback and for assessment of closure. The articulation studies are to develop and evaluate remedial speech procedures. Response to treatment is predicted from articulation and other measures, and predictions are made through regression procedure. Also, the estimated omega square statistic is used in group comparisons to determine variance accounted for. Study variables include articulation, language, reading, listening, psychosocial, and dental measures. The articulation studies also use clustering procedure to identify homogeneous subject subsets. Remediation is conceptualized with a motor learning model with response acquisition and automatization steps.